E21-1 (a) We’ll assume that the new temperature scale is related to the Celsius scale by a linear

transformation; then TS = mTC + b, where m and b are constants to be determined, TS is the

temperature measurement in the “new” scale, and TC is the temperature measurement in Celsius
degrees.
One of our known points is absolute zero;

TS = mTC + b,
(0) = m(−273.15◦ C) + b.

We have two other points, the melting and boiling points for water,

(TS )bp = m(100◦ C) + b,

(TS )mp = m(0◦ C) + b;

we can subtract the top equation from the bottom equation to get

(TS )bp − (Ts )S mp = 100 C◦ m.

We are told this is 180 S◦ , so m = 1.8 S◦ /C◦ . Put this into the first equation and then find b,
b = 273.15◦ Cm = 491.67◦ S. The conversion is then

TS = (1.8 S◦ /C◦ )TC + (491.67◦ S).

(b) The melting point for water is 491.67◦ S; the boiling point for water is 180 S◦ above this, or
671.67◦ S.

E21-2 T F = 9(−273.15 deg C)/5 + 32◦ F = −459.67◦ F.

E21-3 (a) We’ll assume that the new temperature scale is related to the Celsius scale by a linear
transformation; then TS = mTC + b, where m and b are constants to be determined, TS is the
temperature measurement in the “new” scale, and TC is the temperature measurement in Celsius
degrees.
One of our known points is absolute zero;

TS = mTC + b,
(0) = m(−273.15◦ C) + b.

We have two other points, the melting and boiling points for water,

(TS )bp = m(100◦ C) + b,

(TS )mp = m(0◦ C) + b;

we can subtract the top equation from the bottom equation to get

(TS )bp − (Ts )S mp = 100 C◦ m.

We are told this is 100 Q◦ , so m = 1.0 Q◦ /C◦ . Put this into the first equation and then find b,
b = 273.15◦ C = 273.15◦ Q. The conversion is then

TS = TC + (273.15◦ S).

(b) The melting point for water is 273.15◦ Q; the boiling point for water is 100 Q◦ above this, or
373.15◦ Q.
(c) Kelvin Scale.

265
E21-4 (a) T = (9/5)(6000 K − 273.15) + 32 = 10000◦ F.
(b) T = (5/9)(98.6◦ F − 32) = 37.0◦ C.
(c) T = (5/9)(−70◦ F − 32) = −57◦ C.
(d) T = (9/5)(−183◦ C) + 32 = −297◦ F.
(e) It depends on what you think is hot. My mom thinks 79◦ F is too warm; that’s T =
(5/9)(79◦ F − 32) = 26◦ C.

E21-5 T = (9/5)(310 K − 273.15) + 32 = 98.3◦ F, which is fine.

E21-6 (a) T = 2(5/9)(T − 32), so −T /10 = −32, or T = 320◦ F.

(b) 2T = (5/9)(T − 32), so 13T /5 = −32, or T = −12.3◦ F.

E21-7 If the temperature (in Kelvin) is directly proportional to the resistance then T = kR,
where k is a constant of proportionality. We are given one point, T = 273.16 K when R = 90.35 Ω,
but that is okay; we only have one unknown, k. Then (273.16 K) = k(90.35 Ω) or k = 3.023 K/Ω.
If the resistance is measured to be R = 96.28 Ω, we have a temperature of

T = kR = (3.023 K/Ω)(96.28 Ω) = 291.1 K.

E21-8 T = (510◦ C)/(0.028 V)V , so T = (1.82×104◦ C/V)(0.0102 V) = 186◦ C.

E21-9 We must first find the equation which relates gain to temperature, and then find the gain
at the specified temperature. If we let G be the gain we can write this linear relationship as

G = mT + b,

where m and b are constants to be determined. We have two known points:

(30.0) = m(20.0◦ C) + b,
(35.2) = m(55.0◦ C) + b.

If we subtract the top equation from the bottom we get 5.2 = m(35.0◦ C), or m = 1.49 C−1 . Put
this into either of the first two equations and

(30.0) = (0.149 C−1 )(20.0◦ C) + b,

which has a solution b = 27.0

Now to find the gain when T = 28.0 ◦ C:

G = mT + b = (0.149 C−1 )(28.0◦ C) + (27.0) = 31.2

E21-10 p/ptr = (373.15 K)/(273.16 K) = 1.366.

E21-11 100 cm Hg is 1000 torr. P He = (100 cm Hg)(373 K)/(273.16 K) = 136.550 cm Hg. Ni-
trogen records a temperature which is 0.2 K higher, so P N = (100 cm Hg)(373.2 K)/(273.16 K) =
136.623 cm Hg. The difference is 0.073 cm Hg.

E21-12 ∆L = (23×10−6 /C◦ )(33 m)(15C◦ ) = 1.1×10−2 m.

E21-13 ∆L = (3.2×10−6 /C◦ )(200 in)(60C◦ ) = 3.8×10−2 in.

266
E21-14 L0 = (2.725cm)[1 + (23×10−6 /C◦ )(128C◦ )] = 2.733 cm.

E21-15 We want to focus on the temperature change, not the absolute temperature. In this
case, ∆T = T f − T i = (42◦ C) − (−5.0◦ C) = 47 C◦ .
Then
∆L = (11 × 10−6 C−1 )(12.0 m)(47 C◦ ) = 6.2 × 10−3 m.

E21-16 ∆A = 2αA∆T , so
∆A = 2(9×10−6 /C◦ )(2.0 m)(3.0 m)(30C◦ ) = 3.2×10−3 m2 .

E21-17 (a) We’ll apply Eq. 21-10. The surface area of a cube is six times the area of one face,
which is the edge length squared. So A = 6(0.332 m)2 = 0.661 m2 . The temperature change is
∆T = (75.0◦ C) − (20.0◦ C) = 55.0 C◦ . Then the increase in surface area is
∆A = 2αA∆T = 2(19 × 10−6 C−1 )(0.661 m2 )(55.0 C◦ ) = 1.38 × 10−3 m2
(b) We’ll now apply Eq. 21-11. The volume of the cube is the edge length cubed, so
V = (0.332 m)3 = 0.0366 m3 .
and then from Eq. 21-11,
∆V = 2αV ∆T = 3(19 × 10−6 C−1 )(0.0366 m3 )(55.0 C◦ ) = 1.15 × 10−4 m3 ,
is the change in volume of the cube.

E21-18 V 0 = V (1 + 3α∆T ), so
V 0 = (530 cm3 )[1 + 3(29×10−6 /C◦ )(−172 C◦ )] = 522 cm3 .

E21-19 (a) The slope is approximately 1.6×10−4 /C◦ .

(b) The slope is zero.

E21-20 ∆r = (β/3)r∆T , so
∆r = [(3.2×10−5 /K)/3](6.37×106 m)(2700 K) = 1.8×105 m.

E21-21 We’ll assume that the steel ruler measures length correctly at room temperature. Then
the 20.05 cm measurement of the rod is correct. But both the rod and the ruler will expand in the
oven, so the 20.11 cm measurement of the rod is not the actual length of the rod in the oven. What
is the actual length of the rod in the oven? We can only answer that after figuring out how the 20.11
cm mark on the ruler moves when the ruler expands.
Let L = 20.11 cm correspond to the ruler mark at room temperature. Then
∆L = αsteel L∆T = (11 × 10−6 C−1 )(20.11 cm)(250 C◦ ) = 5.5 × 10−2 cm
is the shift in position of the mark as the ruler is raised to the higher temperature. Then the change
in length of the rod is not (20.11 cm) − (20.05 cm) = 0.06 cm, because the 20.11 cm mark is shifted
out. We need to add 0.055 cm to this; the rod changed length by 0.115 cm.
The coefficient of thermal expansion for the rod is
∆L (0.115 cm)
α= = = 23 × 10−6 C−1 .
L∆T (20.05 cm)(250 C◦ )

267
E21-22 A = ab, A0 = (a + ∆a)(b + ∆b) = ab + a∆b + b∆a + ∆a∆b, so

∆A = a∆b + b∆a + ∆a∆b,

= A(∆b/b + ∆a/a + ∆a∆b/ab),
≈ A(α∆T + α∆T ),
= 2αA∆T.

E21-23 Solve this problem by assuming the solid is in the form of a cube.
If the length of one side of a cube is originally L0 , then the volume is originally V0 = L30 . After
heating, the volume of the cube will be V = L3 , where L = L0 + ∆L.
Then

V = L3 ,
= (L0 + ∆L)3 ,
= (L0 + αL0 ∆T )3 ,
= L30 (1 + α∆T )3 .

As long as the quantity α∆T is much less than one we can expand the last line in a binomial
expansion as
V ≈ V0 (1 + 3α∆T + · · ·),
so the change in volume is ∆V ≈ 3αV0 ∆T .

E21-24 (a) ∆A/A = 2(0.18%) = (0.36%).

(b) ∆L/L = 0.18%.
(c) ∆V /V = 3(0.18%) = (0.54%).
(d) Zero.
(e) α = (0.0018)/(100 C◦ ) = 1.8×10−5 /C◦ .

E21-25 ρ0 − ρ = m/V 0 − m/V = m/(V + ∆V ) − m/V ≈ −m∆V /V 2 . Then

∆ρ = −(m/V )(∆V /V ) = −ρβ∆T.

E21-26 Use the results of Exercise 21-25.

(a) ∆V /V = 3∆L/L = 3(0.092%) = 0.276%. The change in density is

∆ρ/ρ = −∆V /V = −(0.276%) = −0.28

(b) α = β/3 = (0.28%)/3(40 C◦ ) = 2.3×10−5 /C◦ . Must be aluminum.

E21-27 The diameter of the rod as a function of temperature is

ds = ds,0 (1 + αs ∆T ),

The diameter of the ring as a function of temperature is

db = db,0 (1 + αb ∆T ).

268
We are interested in the temperature when the diameters are equal,

ds,0 (1 + αs ∆T ) = db,0 (1 + αb ∆T ),
αs ds,0 ∆T − αb db,0 ∆T = db,0 − ds,0 ,
db,0 − ds,0
∆T = ,
αs ds,0 − αb db,0
(2.992 cm) − (3.000 cm)
∆T = ,
(11×10 /C )(3.000 cm) − (19×10−6 /C◦ )(2.992 cm)
−6 ◦

= 335 C◦ .

The final temperature is then T f = (25◦ ) + 335 C◦ = 360◦ .

E21-28 (a) ∆L = ∆L1 + ∆L2 = (L1 α1 + L2 α2 )∆T . The effective value for α is then

∆L α1 L1 + α2 L2
α= = .
L∆T L
(b) Since L2 = L − L1 we can write

α1 L1 + α2 (L − L1 ) = αL,
α − α2
L1 = L ,
α1 − α2
(13×10−6 ) − (11×10−6 )
= (0.524 m) = 0.131 m.
(19×10−6 ) − (11×10−6 )

The brass length is then 13.1 cm and the steel is 39.3 cm.

E21-29 At 100◦ C the glass and mercury each have a volume V0 . After cooling, the difference in
volume changes is given by
∆V = V0 (3αg − β m )∆T.
Since m = ρV , the mass of mercury that needs to be added can be found by multiplying though by
the density of mercury. Then

∆m = (0.891 kg)[3(9.0×10−6 /C◦ ) − (1.8×10−4 /C◦ )](−135C◦ ) = 0.0184 kg.

This is the additional amount required, so the total is now 909 g.

E21-30 (a) The rotational inertia is given by I = r2 dm; changing the temperature requires
R

r → r0 = r + ∆r = r(1 + α∆T ). Then

Z Z
I 0 = (1 + α∆T )2 r2 dm ≈ (1 + 2α∆T ) r2 dm,

so ∆I = 2αI∆T .
(b) Since L = Iω, then 0 = ω∆I + I∆ω. Rearranging, ∆ω/ω = −∆I/I = −2α∆T . Then

∆ω = −2(19×10−6 /C◦ )(230 rev/s)(170 C◦ ) = −1.5 rev/s.

269
 E21-31 This problem is related to objects which expand when heated, but we never actually need
to calculate any temperature changes. We will, however, be interested in the change in rotational
inertia. Rotational inertia is directly proportional to the square of the (appropriate) linear dimension,
so
I f /I i = (rf /ri )2 .
(a) If the bearings are frictionless then there are no external torques, so the angular momentum
is constant.
(b) If the angular momentum is constant, then

Li = Lf ,
I i ωi = I f ωf .

We are interested in the percent change in the angular velocity, which is

 2  2
ωf − ωi ωf Ii ri 1
= −1= −1= −1= − 1 = −0.36%.
ωi ωi If rf 1.0018

(c) The rotational kinetic energy is proportional to Iω 2 = (Iω)ω = Lω, but L is constant, so
Kf − Ki ωf − ωi
= = −0.36%.
Ki ωi

E21-32 (a) The period of a physical pendulum is given by Eq. 17-28. There are two variables
in the equation that depend on length. I, which is proportional to a length squared, and d, which
is proportional √
to a length. This means that the period have an overall dependence on length
proportional to r. Taking the derivative,
1P 1
∆P ≈ dP = dr ≈ P α∆T.
2r 2
(b) ∆P/P = (0.7×10−6 C◦ )(10C◦ )/2 = 3.5×10−6 . After 30 days the clock will be slow by

∆t = (30 × 24 × 60 × 60 s)(3.5×10−6 ) = 9.07 s.

E21-33 Refer to the Exercise 21-32.

∆P = (3600 s)(19×10−6 C◦ )(−20C◦ )/2 = 0.68 s.

E21-34 At 22◦ C the aluminum cup and glycerin each have a volume V0 . After heating, the
difference in volume changes is given by

∆V = V0 (3αa − β g )∆T.

The amount that spills out is then

∆V = (110 cm3 )[3(23×10−6 /C◦ ) − (5.1×10−4 /C◦ )](6C◦ ) = −0.29 cm3 .

E21-35 At 20.0◦ C the glass tube is filled with liquid to a volume V0 . After heating, the difference
in volume changes is given by
∆V = V0 (3αg − β l )∆T.
The cross sectional area of the tube changes according to

∆A = A0 2αg ∆T.

270
Consequently, the height of the liquid changes according to

∆V = (h0 + ∆h)(A0 + ∆A) − h0 A,

≈ h0 ∆A + A0 ∆h,
∆V /V0 = ∆A/A0 + ∆h/h0 .

Then
∆h = (1.28 m/2)[(1.1×10−5 /C◦ ) − (4.2×10−5 /C◦ )](13 C◦ ) = 2.6×10−4 m.

E21-36 (a) β = (dV /dT )/V . If pV = nRT , then p dV = nR dT , so

β = (nR/p)/V = nR/pV = 1/T.

(b) Kelvins.
(c) β ≈ 1/(300/K) = 3.3×10−3 /K.

E21-37 (a) V = (1 mol)(8.31 J/mol · K)(273 K)/(1.01×105 Pa) = 2.25×10−2 m3 .

(b) (6.02×1023 mol−1 )/(2.25×104 /cm3 ) = 2.68×1019 .

E21-38 n/V = p/kT , so

3
n/V = (1.01×10−13 Pa)/(1.38×10−23 J/K)(295 K) = 25 part/cm .

E21-39 (a) Using Eq. 21-17,

pV (108×103 Pa)(2.47 m3 )
n= = = 113 mol.
RT (8.31 J/mol·K)([12 + 273] K)

(b) Use the same expression again,

nRT (113 mol)(8.31 J/mol·K)([31 + 273] K)

V = = = 0.903 m3 .
p (316×103 Pa)

E21-40 (a) n = pV /RT = (1.01×105 Pa)(1.13×10−3 m3 )/(8.31 J/mol·K)(315 K) = 4.36×10−2 mol.

(b) T f = T i pf V f /pi V i , so

(315 K)(1.06×105 Pa)(1.530×10−3 m3 )

Tf = = 448 K.
(1.01×105 Pa)(1.130×10−3 m3 )

2 2
E21-41 pi = (14.7 + 24.2) lb/in = 38.9 lb/in . pf = pi T f V i /T i V f , so
2
(38.9 lb/in )(299K)(988 in3 ) 2
pf = = 41.7 lb/in .
(270 K)(1020 in3 )
2 2
The gauge pressure is then (41.7 − 14.7) lb/in = 27.0 lb/in .

E21-42 Since p = F/A and F = mg, a reasonable estimate for the mass of the atmosphere is

m = pA/g = (1.01×105 Pa)4π(6.37×106 m)2 /(9.81 m/s2 ) = 5.25×1018 kg.

271
E21-43 p = p0 + ρgh, where h is the depth. Then P f = 1.01×105 Pa and
pi = (1.01×105 Pa) + (998 kg/m3 )(9.81 m/s2 )(41.5 m) = 5.07×105 Pa.
V f = V i pi T f /pf T i , so
(19.4 cm3 )(5.07×105 Pa.)(296 K)
Vf = = 104 cm3 .
(1.01×105 Pa)(277 K)

E21-44 The new pressure in the pipe is

pf = pi V i /V f = (1.01×105 Pa)(2) = 2.02×105 Pa.
The water pressure at some depth y is given by p = p0 + ρgy, so
(2.02×105 Pa) − (1.01×105 Pa)
y= = 10.3 m.
(998 kg/m3 )(9.81 m/s2 )
Then the water/air interface inside the tube is at a depth of 10.3 m; so h = (10.3 m) + (25.0 m)/2 =
22.8 m.

P21-1 (a) The dimensions of A must be [time]−1 , as can be seen with a quick inspection of the
equation. We would expect that A would depend on the surface area at the very least; however,
that means that it must also depend on some other factor to fix the dimensionality of A.
(b) Rearrange and integrate,
Z T Z t
d∆T
= − A dt,
∆T0 ∆T 0
ln(∆T /∆T0 ) = −At,
∆T = ∆T0 e−At .

P21-2 First find A.

ln(∆T0 /∆T ) ln[(29 C◦ )/(25 C◦ )]
A= = = 3.30×10−3 /min.
t (45 min)
Then find time to new temperature difference.
ln(∆T0 /∆T ) ln[(29 C◦ )/(21 C◦ )]
t= = = 97.8min
t (3.30×10−3 /min)
This happens 97.8 − 45 = 53 minutes later.

P21-3 If we neglect the expansion of the tube then we can assume the cross sectional area of the
tube is constant. Since V = Ah, we can assume that ∆V = A∆h. Then since ∆V = βV0 ∆T , we
can write ∆h = βh0 ∆T .

P21-4 For either container we can write pi Vi = ni RTi . We are told that Vi and ni are constants.
Then ∆p = AT1 − BT2 , where A and B are constants. When T1 = T2 ∆p = 0, so A = B. When
T1 = T tr and T2 = T b we have
(120 mm Hg) = A(373 K − 273.16 K),
so A = 1.202 mm Hg/K. Then
(90 mm Hg) + (1.202 mm Hg/K)(273.16 K)
T = = 348 K.
(1.202 mm Hg/K)
Actually, we could have assumed A was negative, and then the answer would be 198 K.

272
P21-5 Start with a differential form for Eq. 21-8, dL/dT = αL0 , rearrange, and integrate:
Z L Z T
dL = αL0 dT,
L0 T0
Z T
L − L0 = L0 α dT,
T0
!
Z T
L = L0 1+ α dT .
T0

P21-6 ∆L = αL∆T , so
∆T 1 ∆L (96×10−9 m/s)
= = = 0.23◦ C/s.
∆t αL ∆t (23×10−6 /C◦ )(1.8×10−2 m)

P21-7 (a) Consider the work that was done for Ex. 21-27. The length of rod a is
La = La,0 (1 + αa ∆T ),
while the length of rod b is
Lb = Lb,0 (1 + αb ∆T ).
The difference is
La − Lb = La,0 (1 + αa ∆T ) − Lb,0 (1 + αb ∆T ),
= La,0 − Lb,0 + (La,0 αa − Lb,0 αb )∆T,
which will be a constant is La,0 αa = Lb,0 αb or
Li,0 ∝ 1/αi .
(b) We want La,0 − Lb,0 = 0.30 m so
k/αa − k/αb = 0.30 m,
where k is a constant of proportionality;
k = (0.30 m)/ 1/(11×10−6 /C◦ ) − 1/(19×10−6 /C◦ ) = 7.84×10−6 m/C◦ .

The two lengths are

La = (7.84×10−6 m/C◦ )/(11×10−6 /C◦ ) = 0.713 m
for steel and
Lb = (7.84×10−6 m/C◦ )/(19×10−6 /C◦ ) = 0.413 m
for brass.

P21-8 The fractional increase in length of the bar is ∆L/L0 = α∆T. The right triangle on the left
has base L0 /2, height x, and hypotenuse (L0 + ∆L)/2. Then
r
1 L ∆L
q
2 0
x= (L0 + ∆L)2 − L0 = 2 .
2 2 L0
With numbers,
(3.77 m) p
x= 2(25×10−6 /C◦ )(32 C◦ ) = 7.54×10−2 m.
2

273
 R
P21-9 We want to evaluate V = V0 (1 + β dT ); the integral is the area under the graph; the
graph looks like a triangle, so the result is
V = V0 [1 + (16 C◦ )(0.0002/C◦ )/2] = (1.0016)V0 .
The density is then
ρ = ρ0 (V0 /V ) = (1000 kg/m3 )/(1.0016) = 0.9984 kg/m3 .

P21-10 At 0.00◦ C the glass bulb is filled with mercury to a volume V0 . After heating, the difference
in volume changes is given by
∆V = V0 (β − 3α)∆T.
Since T0 = 0.0◦ C, then ∆T = T , if it is measured in ◦ C. The amount of mercury in the capillary is
∆V , and since the cross sectional area is fixed at A, then the length is L = ∆V /A, or
V
L= (β − 3α)∆T.
A

P21-11 Let a, b, and c correspond to aluminum, steel, and invar, respectively. Then
a2 + b2 − c2
cos C = .
2ab
We can replace a with a0 (1 + αa ∆T ), and write similar expressions for b and c. Since a0 = b0 = c0 ,
this can be simplified to
(1 + αa ∆T )2 + (1 + αb ∆T )2 − (1 + αc ∆T )2
cos C = .
2(1 + αa ∆T )(1 + αb ∆T )
Expand this as a Taylor series in terms of ∆T , and we find
1 1
cos C ≈ + (αa + αb − 2αc ) ∆T.
2 2
Now solve:
2 cos(59.95◦ ) − 1
∆T = = 46.4◦ C.
(23×10−6 /C◦ ) + (11×10−6 /C◦ ) − 2(0.7×10−6 /C◦ )
The final temperature is then 66.4◦ C.

P21-12 The bottom of the iron bar moves downward according to ∆L = αL∆T . The center of
mass of the iron bar is located in the center; it moves downward half the distance. The mercury
expands in the glass upwards; subtracting off the distance the iron moves we get
∆h = βh∆T − ∆L = (βh − αL)∆T.
The center of mass in the mercury is located in the center. If the center of mass of the system is to
remain constant we require
mi ∆L/2 = mm (∆h − ∆L)/2;
or, since ρ = mV = mAy,
ρi αL = ρm (βh − 2αL).
Solving for h,
(12×10−6 /C◦ )(1.00 m)[(7.87×103 kg/m3 ) + 2(13.6×103 kg/m3 )]
h= = 0.17 m.
(13.6×103 kg/m3 )(18×10−5 /C◦ )

274
 P21-13 The volume of the block which is beneath the surface of the mercury displaces a mass
of mercury equal to the mass of the block. The mass of the block is independent of the temperature
but the volume of the displaced mercury changes according to

V m = V m,0 (1 + β m ∆T ).

This volume is equal to the depth which the block sinks times the cross sectional area of the block
(which does change with temperature). Then

hs hb 2 = hs,0 hb,0 2 (1 + β m ∆T ),

where hs is the depth to which the block sinks and hb,0 = 20 cm is the length of the side of the
block. But
hb = hb,0 (1 + αb ∆T ),
so
1 + β m ∆T
hs = hs,0 .
(1 + αb ∆T )2
Since the changes are small we can expand the right hand side using the binomial expansion; keeping
terms only in ∆T we get
hs ≈ hs,0 (1 + (β m − 2αb )∆T ),
which means the block will sink a distance hs − hs,0 given by

hs,0 (β m − 2αb )∆T = hs,0 (1.8×10−4 /C◦ ) − 2(23×10−6 /C◦ ) (50 C◦ ) = (6.7×10−3 )hs,0 .
 

In order to finish we need to know how much of the block was submerged in the first place. Since
the fraction submerged is equal to the ratio of the densities, we have

hs,0 /hb,0 = ρb /ρm = (2.7×103 kg/m3 )/(1.36×104 kg/m3 ),

so hs,0 = 3.97 cm, and the change in depth is 0.27 mm.

P21-14 The area of glass expands according to ∆Ag = 2αg Ag ∆T . The are of Dumet wire expands
according to
∆Ac + ∆Ai = 2(αc Ac + αi Ai )∆T.
We need these to be equal, so

αg Ag = αc Ac + αi Ai ,
αg rg 2 = αc (rc 2 − ri 2 ) + αi ri 2 ,
αg (rc 2 + ri 2 ) = αc (rc 2 − ri 2 ) + αi ri 2 ,
ri 2 αc − αg
= .
rc 2 αc − αi

P21-15

P21-16 V2 = V1 (p1 /p2 )(T1 /T2 ), so

V2 = (3.47 m3 )[(76 cm Hg)/(36 cm Hg)][(225 K)/(295 K)] = 5.59 m3 .

275
 P21-17 Call the containers one and two so that V1 = 1.22 L and V2 = 3.18 L. Then the initial
number of moles in the two containers are
pi V 1 pi V 2
n1,i = and n2,i = .
RT i RT i
The total is
n = pi (V1 + V2 )/(RT i ).
Later the temperatures are changed and then the number of moles of gas in each container is
pf V 1 pf V 2
n1,f = and n2,f = .
RT 1,f RT 2,f

The total is still n, so  

pf V1 V2 pi (V1 + V2 )
+ = .
R T 1,f T 2,f RT i
We can solve this for the final pressure, so long as we remember to convert all temperatures to
Kelvins,
 −1
pi (V1 + V2 ) V1 V2
pf = + ,
Ti T 1,f T 2,f
or  −1
(1.44 atm)(1.22L + 3.18 L) (1.22 L) (3.18 L)
pf = + = 1.74 atm.
(289 K) (289 K) (381 K)

P21-18 Originally nA = pA VA /RTA and nB = pB VB /RTB ; VB = 4VA . Label the final state of
A as C and the final state of B as D. After mixing, nC = pC VA /RTA and nD = pD VB /RTB , but
PC = PD and nA + nB = nC + nD . Then

pA /TA + 4pB /TB = pC (1/TA + 4/TB ),

or
(5×105 Pa)/(300 K) + 4(1×105 Pa)/(400 K)
pC = = 2.00×105 Pa.
1/(300 K) + 4/(400 K)

P21-19 If the temperature is uniform then all that is necessary is to substitute p0 = nRT /V and
p = nRT /V ; cancel RT from both sides, and then equate n/V with nV .

P21-20 Use the results of Problem 15-19. The initial pressure inside the bubble is pi = p0 + 4γ/ri .
The final pressure inside the bell jar is zero, sopf = 4γ/rf . The initial and final pressure inside the
bubble are related by pi ri 3 = pf rf 3 = 4γrf 2 . Now for numbers:

pi = (1.01×105 Pa) + 4(2.5×10−2 N/m)/(2.0×10−3 m) = 1.0105×105 Pa.

and s
(1.0105×105 Pa)(2.0×10−3 m)3
rf = = 8.99×10−2 m.
4(2.5×10−2 N/m)

P21-21

P21-22

276